Persistent immune activation is a strong predictor of disease progression in HIV infection and is characterized by activation of all immune cell types, increased levels of proinflammatory cytokines and activation-induced lymphocyte apoptosis. Translocation of microbial toll-like receptor (TLR) ligands from the gut to the systemic circulatio has been recognized in recent years to be a major factor in inducing immune activation. The resulting increase in inflammatory cytokines may be the fundamental cause for immune activation and determinant of disease progression. The cytokine storm, immune activation and lymphocyte apoptosis attended with microbial products in systemic circulation seen in HIV infection also characterizes natural and experimental sepsis. HMGB1 secreted from macrophages and dendritic cells acts as a master regulator of cytokine storm in sepsis. We have recently shown that silencing HMGB1 in human macrophages and dendritic cells in vivo in humanized BLT mice by targeted delivery of siRNA (via a peptide called RVG-9R) dramatically suppressed the sepsis-induced cytokine storm and mortality. Although serum level of HMGB1 as well as downstream cytokines are also elevated in HIV infected individuals, their role in chronic immune activation is not clear. In this proposal, we will test the role of HMGB1 and downstream cytokines, their relation to microbial products and the effect of suppressing these inflammatory mediators on immune activation and disease progression in vivo using the humanized mouse model for HIV infection. In Aim 1, we will characterize the full spectrum of immune activation in HIV infected humanized mice without or with ART. In the second aim we will test if silencing HMGB1 or downstream cytokines by targeted delivery of siRNA to macrophages and dendritic cells can reduce immune activation and disease progression in infected mice.